AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Block-State Transformers

Jonathan Pilault, Mahan Fathi, Orhan Firat, Christopher Pal, Pierre-Luc Bacon, Ross Goroshin

NeurIPS 2023 · Architectures and efficiency › Attention and transformers · approach: Sparse efficient transformers for long context · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
place single BST layer at index 9 instead of 3 (ablation)BST:SH:S4 · PG19perplexity: 12.41 -> 11.88Improved
single BST layer at index 12 instead of 9 (ablation)BST:SH:S4 · PG19perplexity: 11.88 -> 12.03Hurt
increase number of BST layers from 2 to 4 (ablation)BST:SH:S4 · PG19perplexity: 11.69 -> 11.21Improved
increase number of BST layers from 4 to 5 (ablation)BST:SH:S4 · PG19perplexity: 11.21 -> 11.20No clear effect
increase S4 state size from 8 to 16 (ablation)BST:SH:S4 · PG19perplexity: 11.95 -> 11.57Improved
increase S4 state size from 16 to 64 (ablation)BST:SH:S4 · PG19perplexity: 11.57 -> 11.54No clear effect
use BST:MF instead of BST:SH (S4)BST:SH:S4 · arXivperplexity: 2.51 -> 2.48Improved
use BST:SH instead of BST:MF (S4)BST:SH:S4 · PG19perplexity: 11.63 -> 11.57Improved
use unstructured filters instead of S4 (SH)BST:SH:S4 · PG19perplexity: 11.57 -> 11.52Improved
BST:SH:S4 versus Mega-chunkBST:SH:S4 · LRA Avgaccuracy: 85.66 -> 86.96Improved
BST:SH:S4 versus BRecTBST:SH:S4 · LRA Avgaccuracy: 60.80 -> 86.96Improved
BST:SH:S4-L versus GSS-Hybrid-LBST:SH:S4 · PG19perplexity: 10.52 -> 10.47Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy